general ®electric...v. anselmo 1 study management cal space j. arnold \ university/ industry...
TRANSCRIPT
![Page 1: GENERAL ®ELECTRIC...V. ANSELMO 1 STUDY MANAGEMENT CAL SPACE J. ARNOLD \ UNIVERSITY/ INDUSTRY COMMITTEE SENIOR TECHNICAL ^ COMMITTEE DR. FROSCH, CHAIRMAN NASA _^ AUTOMATION _ …](https://reader035.vdocuments.mx/reader035/viewer/2022071412/610974a6dd4dbd69785f4e76/html5/thumbnails/1.jpg)
" '<'-" ;*,.,.
CONTRACT NO. NAS 5-25182P.O. 167-HA 3463
9 JULY 1984TASK ASSIGNMENT NR 740-9
- FINAL REPORT -
SPACE STATION AUTOMATION STUDY
AUTOMATION REQUIREMENTS DERIVED FROMSPACE MANUFACTURING CONCEPTS
(NASA-CB-177862-Vcl-l) SPACE STATION N86-27399AUTOMATION STUDY:' AUTCMAIICN EEQUIBEMENTSDERIVED FEOM SPACE MANUFACTUEING CONCEPTS. ,VOLUME 1: EXECUTIVE SUMMABY Final Report Unclas
Co. ) 34 p_EC A03/MF A P I . 63/18 .4.3115
VOLUME IEXECUTIVE SUMMARY
SPACE SYSTEMS DIVISION
Valley Forge Space CenterP.O. Box 8555 Philadelphia, PA 19101
G E N E R A L ®ELECTRIC
![Page 2: GENERAL ®ELECTRIC...V. ANSELMO 1 STUDY MANAGEMENT CAL SPACE J. ARNOLD \ UNIVERSITY/ INDUSTRY COMMITTEE SENIOR TECHNICAL ^ COMMITTEE DR. FROSCH, CHAIRMAN NASA _^ AUTOMATION _ …](https://reader035.vdocuments.mx/reader035/viewer/2022071412/610974a6dd4dbd69785f4e76/html5/thumbnails/2.jpg)
CONTENTS
1.0 Introduction 1
2.0 Study Objectives, Guidelines and Approach 6
3.0 Study Results 8
3.1 GaAs Electroepitaxial Crystal Production 9
and Wafer Manufacturing Facility Description
3.2 Crystal Production and Wafer Manufacturing 13
Facility Automation Requirements
3.3 Microelectronics Chip Processing Facility 16
Description
3.4 GaAs Microelectronics Chip Processing 21
Facility Automation Requirements
4.0 Conclusions 24
5.0 Recommendations 29
![Page 3: GENERAL ®ELECTRIC...V. ANSELMO 1 STUDY MANAGEMENT CAL SPACE J. ARNOLD \ UNIVERSITY/ INDUSTRY COMMITTEE SENIOR TECHNICAL ^ COMMITTEE DR. FROSCH, CHAIRMAN NASA _^ AUTOMATION _ …](https://reader035.vdocuments.mx/reader035/viewer/2022071412/610974a6dd4dbd69785f4e76/html5/thumbnails/3.jpg)
1.0 INTRODUCTION
The purpose of the Space Station Automation Study is
to develop informed technical guidance to NASA in the use
of autonomy and autonomous systems to implement space
station functions.
The study organization is shown in Figure 1.0-1.
NASA headquarters formed and convened a panel of recognized
expert technologists in Automation, Space Science and
Aerospace Engineering. CAL SPACE was assigned the
responsibility for study management, and for convening and
directing a University/Industry Committee to produce the
Space Station Automation Plan. A Senior Technical
Committee, chaired by Dr. Robert Frosch, was appointed to
provide top level technical guidance.
STUDY DIRECTIONNASA HDQTRS
D. HERMANV. ANSELMO
1
STUDY MANAGEMENTCAL SPACE
J. ARNOLD
\
UNIVERSITY/INDUSTRYCOMMITTEE
SENIOR TECHNICAL^ COMMITTEE
DR. FROSCH,CHAIRMAN
NASA_^ AUTOMATION _ EVALUATION
PLAN GROUPA.COHEN
1
• » SUPPORTING STUDIES
TECHNOLOGY ?ffSDESIGN [ TRW - SERVICINGTEAM TEAM 1 HAC - CREW PRODUCTIVITYSRI , ._,.._, . < MMC- CONSTRUCTIONH LUM L.POWELL GE - MANUFACTURING
R. HOOK | BAC-MAN.MArHINP
T f
1 1
COMPARATIVEASSESSMENT
1
LEVELS_. SPACE
STATION. N. HUTCHINSON
Figure 1.0-1. Space Station Automation Study Organization
-1-
![Page 4: GENERAL ®ELECTRIC...V. ANSELMO 1 STUDY MANAGEMENT CAL SPACE J. ARNOLD \ UNIVERSITY/ INDUSTRY COMMITTEE SENIOR TECHNICAL ^ COMMITTEE DR. FROSCH, CHAIRMAN NASA _^ AUTOMATION _ …](https://reader035.vdocuments.mx/reader035/viewer/2022071412/610974a6dd4dbd69785f4e76/html5/thumbnails/4.jpg)
SRI International was assigned to produce quality
focused technology forecasts supporting panel analyses and
guiding system concept design.
A NASA Design Team was convened to study the automation
of remote space operations to produce innovative,
technologically advanced automation concepts and system
designs which will strengthen NASA understanding of
practical autonomy and autonomous systems. Five Aerospace
Contractors, TRW, GE, HAG, MMC and BAG, were assigned to
this team.
The General Electric Company was assigned to assess
automation technology required for remote operations,
including manufacturing applications. In carrying out this
assignment, GE assessed over one hundred potential Space
Station missions through an extensive review of proposed
Space Station experiments and manufacturing concepts.
Subsequent meetings of the NASA Design Team resulted in the
direction to proceed with in-depth development of
automation requirements for two manufacturing design
concepts:
(1) Gallium Arsenide Electroepitaxial Crystal
Production and Wafer Manufacturing Facility
(2) Gallium Arsenide VLSI Microelectronics Chip
Processing Facility
Figure 1.0-2 provides a functional overview of the
ultimate design concept incorporating the two manufacturing
facilities on space station. For the purpose of this
study, the concepts were studied separately. This
separation allowed conclusions and results to be determined
in independent time frames without dependent cross ties.
-2-
![Page 5: GENERAL ®ELECTRIC...V. ANSELMO 1 STUDY MANAGEMENT CAL SPACE J. ARNOLD \ UNIVERSITY/ INDUSTRY COMMITTEE SENIOR TECHNICAL ^ COMMITTEE DR. FROSCH, CHAIRMAN NASA _^ AUTOMATION _ …](https://reader035.vdocuments.mx/reader035/viewer/2022071412/610974a6dd4dbd69785f4e76/html5/thumbnails/5.jpg)
GaAs CRYSTAL PRODUCTION ANDWAFER MANUFACTURING
FACILITY
.LABORATORYMODULE
(PRESSURIZED)
I I I fPROCESSING SUBSYSTEMS
J ITELEOPERATORS/ROBOTIC:
J I
ORS/ROBOTICS
ESSING SUBSYSTEPROCESSING SUBS'
I I
MICROELECTRONICS CHIP PROCESSINGFACILITY
Figure 1.0-2 Overview Of The GaAs Manufacturing
Facilities Concepts
Each facility would be developed in an evolutionary
step-by-step process. As they are developed, more and more
automation would be incorporated, evolving towards a full
automation, including maintenance, repair and refurbishment
functions.
Ultimately, in the year 2000 + time frame, it would be
logical that both facilities could be mated to a common,
standard space station pressurized laboratory module. The
part time crew would tend the two facilities from the
laboratory module, where all computer functions of process
control and data display would be performed, and quality
control checks and management of the finished products
accomplished. Either or both facilities could be operated
remotely from the Space Station, however, on separately
powered, unmanned free-flying or tethered platforms, with
control and data flow accomplished by RF communications
with the Space Station or with ground facilities.
Both manufacturing facilities would be contained in
enclosed structures as shown to help manage waste products
and contamination, and to facilitate man-tended repairs and
equipment upgrades.
-3-
![Page 6: GENERAL ®ELECTRIC...V. ANSELMO 1 STUDY MANAGEMENT CAL SPACE J. ARNOLD \ UNIVERSITY/ INDUSTRY COMMITTEE SENIOR TECHNICAL ^ COMMITTEE DR. FROSCH, CHAIRMAN NASA _^ AUTOMATION _ …](https://reader035.vdocuments.mx/reader035/viewer/2022071412/610974a6dd4dbd69785f4e76/html5/thumbnails/6.jpg)
, The electroepitaxial process grows crystals in a low
temperature furnace into ingots. These ingots, typically
three to five inches in diameter, are then sliced into very
thin wafers. The process provides defect free Gallium
Arsenide (GaAs) wafers when accomplished in a gravity-free
environment. In the second concept, many Very Large Scale
Integrated (VLSI) circuits (chips) are typically processed
on each wafer at the same time.
The two concepts were chosen for the main reason that
they both require a very, high degree of automation, and
therefore involve extensive use of teleoperators, robotics,
process mechanization, and artificial intelligence. They
cover both a raw material process and a sophisticated
multi-step process and are therefore highly representative
of the kinds of difficult operation, maintenance, and
repair challenges which can be expected for any type of
space manufacturing facility. The automation techniques
which would be developed for these space missions will
provide . direct benefits in the design of future
ground-based automated factories to be used for a wide
variety of materials processing and manufacturing
applications.
Supporting reasons for selecting the two concepts are:
(1) There is a growing demand for faster, larger, and
radiation hardened Integrated Circuits for which
. Gallium Arsenide has superior characteristics over
silicon.
(2.) An ultra-clean environment is necessary for
efficient electroepitaxial crystal growth (EGG)
and manufacturing of GaAs products. Additionally
EGG requires a microgravity environment. Onearth, EGG can only grow crystals of small size
and value because of gray.ity-induced convection
currents.
-4-
![Page 7: GENERAL ®ELECTRIC...V. ANSELMO 1 STUDY MANAGEMENT CAL SPACE J. ARNOLD \ UNIVERSITY/ INDUSTRY COMMITTEE SENIOR TECHNICAL ^ COMMITTEE DR. FROSCH, CHAIRMAN NASA _^ AUTOMATION _ …](https://reader035.vdocuments.mx/reader035/viewer/2022071412/610974a6dd4dbd69785f4e76/html5/thumbnails/7.jpg)
(3) The two concepts are compatible with each" other.
Although the Crystal Production/Wafer Manufacturing
Facility could probably be flown five years before the
Microelectronics Chip Processing Facility, eventually
the product of one would provide the wafers- to be
processed into chips by the other.
The study results, although specifically addressing crystal
growth and chip production, identify generic areas which will
require significant further study for any planned future
manufacturing in space. While cost analysis is beyond the scope
of this report, the economics and benefits of any space
manufacturing facility must be closely analyzed. The success of
Space Station will be determined to a large extent by the
programs ability to stimulate development of advanced
technologies and fully develop the commercial potential of space.
Advanced technologies for the automation of maintenance, repair,
and refurbishment activities, as well as contamination control
and waste removal represent major technological challenges to 'any
space based manufacturing facility. Advanced designs of space
manufacturing facilities will employ a high degree of automation,
however the initial designs will be based on state-of-the-art
hard automation such as terrestrial factories are employing and
will grow and evolve as space and terrestrial technologies
mature. > • '
The unique aspects of a space manufacturing facility compared
to a similar terrestrial factory, include the inability to bring
in technicians and specialists for maintenance and malfunction
repair. Therefore the advanced automation technology
requirements identified by the study are those systems required
to remotely monitor, diagnose, and automatically reconfigure,
maint'ain and repair in the event of malfunction. These
requirements embrace a broad spectrum of enabling technologies
ranging from ultimate expert systems for monitoring, diagnosis
and reconfiguration to teleoperation and robotic manipulative
systems to perform manufacturing, servicing and repair under
remote control from either the Space Station or ground.
-5-
![Page 8: GENERAL ®ELECTRIC...V. ANSELMO 1 STUDY MANAGEMENT CAL SPACE J. ARNOLD \ UNIVERSITY/ INDUSTRY COMMITTEE SENIOR TECHNICAL ^ COMMITTEE DR. FROSCH, CHAIRMAN NASA _^ AUTOMATION _ …](https://reader035.vdocuments.mx/reader035/viewer/2022071412/610974a6dd4dbd69785f4e76/html5/thumbnails/8.jpg)
c 2.0 STUDY OBJECTIVES, GUIDELINES AND APPROACH
The GE portion of the study was led by the Space
Systems Division, but also utilized the corporate
experience in manufacturing and automation in other GE
divisions. The GE work plan is shown in Figure 2.0-1.
TASKJ
REVIEW PREVIOUS STUDIES
TASK;
TASK REQUIREMENT SUMMARY
1
TASK 3
ADVANCED AUTOMATION ASSESSMENT
TASK 4
AUTOMATION PLAN
REVIEW SPACESTATIONMISSIONS
REVIEW AUTOMATIONLITERATURE. SPACE RELATED. TERRESTRIAL
-»•ASSESS ANDPRIORITIZEMISSIONS"
SELECTVIABLEMISSIONS FORFURTHER STUDY
'DEFINEGENERIC SPACEAUTOMATIONCONCEPTS
~^~*
DEFINE GENERICSUBTASKS
DETERMIN• BILITYOF
MATION C(FOR SUBT>
-»FOCUS ONTWO -HMANUFACTURINGMISSIONS
_ J
E AWL 1C AAUTO-JNCEPTSkSKS
DEVELOP BASELINE„ CONCEPT
• AUTOMATION• ROBOTICS• SENSORS• Al
*
DEVELOP AUTO-MATIONREQUIREMENTSFOR BASELINECONCEPT
GE CORPORATE.SUBCONTRACTOR. AVENDOR SUPPORT
SRI. GE CORPORATESUBCONTRACTOR &/ENDOR SUPPORT
REDIRECTEDORldlTOpr
ASSESSAUTOMATIONCONCEPTS CONCEPTS
&Ui~
DEVELOPAUTOMATIONPLAN ANDRECOMMENDATIOI
DEVELOPAUTOMATIONPLAN
Figure 2.0-1 GE Space Station Automation Study Work Plan
One hundred candidate missions published in the NASA
May 1984 Space Station Mission Requirements Report were
evaluated on the basis of automation potential and
availability of meaningful knowledge. Numerous reports
and technical papers presented in symposia and workshops
and as an outgrowth of funded studies were also reviewed.
As a result of this review the two concepts defined on page
2 were chosen for further development. In order to define
automation requirements for these concepts extensive
-6-
![Page 9: GENERAL ®ELECTRIC...V. ANSELMO 1 STUDY MANAGEMENT CAL SPACE J. ARNOLD \ UNIVERSITY/ INDUSTRY COMMITTEE SENIOR TECHNICAL ^ COMMITTEE DR. FROSCH, CHAIRMAN NASA _^ AUTOMATION _ …](https://reader035.vdocuments.mx/reader035/viewer/2022071412/610974a6dd4dbd69785f4e76/html5/thumbnails/9.jpg)
knowledge of the current manufacturing processes and
development of space based designs was necessary.
Data was provided by Microgravity Research Associates
(MRA) on the GaAs Electroepitaxial Crystal Growth (ECG)
experiment production unit planned for seven STS missions.
MRA assisted GE in developing a baseline concept for the
GaAs Crystal Production/Wafer Manufacturing Facility for
Space Station.
An in-depth analysis of GaAs microelectronics chip
production requirements was developed through evaluation of
GE Microelectronics Processing Facilities and work with Dr.
Keith Russell at the GE Microelectronics Center. The GE
Electronics Laboratory in Syracuse, NY was also evaluated
and provided data and background information on the GaAs
VLSI manufacturing process and the Molecular Beam Epitaxy
experimental laboratory.
Manufacturers of microelectronics processing equipment
were contacted and asked for support. VARIAN, APPLIED
MATERIALS, PERKIN-ELMER, EATON, GCA, ELECTROTECH and HARRIS
all provided valuable literature and information on the
design and operation of commercial processing equipment and
future products. VARIAN visited GE and assisted in the
conceptualization of space based processing equipment
designs. GE engineers visited three manufacturers, VARIAN,
PERKIN-ELMER, APPLIED MATERIALS, to obtain further insight
into process equipment technology for space application.
Design requirements and unconstrained design concepts
were developed for the two missions which consisted of
defined subsystems, facility layouts, and automation
schemes. These were presented at NASA Design Team meetings
and with helpful comments and direction from NASA, SRI, and
the CAL SPACE Automation and Robotics Panel members, a
finalization of the design concepts was undertaken.
— 7 —
![Page 10: GENERAL ®ELECTRIC...V. ANSELMO 1 STUDY MANAGEMENT CAL SPACE J. ARNOLD \ UNIVERSITY/ INDUSTRY COMMITTEE SENIOR TECHNICAL ^ COMMITTEE DR. FROSCH, CHAIRMAN NASA _^ AUTOMATION _ …](https://reader035.vdocuments.mx/reader035/viewer/2022071412/610974a6dd4dbd69785f4e76/html5/thumbnails/10.jpg)
3.0 STUDY RESULTS
The GaAs Crystal Production/Wafer Manufacturing
Facility concept requires a special furnace to provide for
crystal growth into ingots, and equipment for slicing and
polishing the wafers and placing them into cassettes within
containers. Extensive use of robotics and other automation
and mechanization techniques is required for handling of
the raw materials and waste, processing and handling of the
products, and test and servicing functions. The facility
would be highly automated, but man-tended for the purpose
of managing the processes and for maintenance; both ofs
these functions would evolve into nearly totally autonomous
operations through the use of automated servicing and
maintenance functions and control by use of artificial
intelligence concepts once they are fully developed and
proven in space.
The Microelectronics Chip Processing Facility consists
of seven subsystems which are based on latest
state-of-the-art and conceptualized commercial equipment
used for earth-based microelectronics processing. The
terrestrial versions of these subsystems are typically
stand-alone, separate pieces of equipment, sold by various
vendors. Current designs each provide their own computer,
software and handling devices. Wafer loading is usually
accomplished by people in clean rooms using standardized
cassettes each containing about 25 wafers. A vacuum
environment must be accomplished individually by each
subsystem during most steps of microelectronics processing.
Functions of each subsystem were studied and an
evolution into a space version developed. The vacuum
provided by space allows a major simplification of all
subsystems because the vacuum equipment associated with
each subsystem can be eliminated. The resulting ease of
equipment access also permits a very high degree of
-8-
![Page 11: GENERAL ®ELECTRIC...V. ANSELMO 1 STUDY MANAGEMENT CAL SPACE J. ARNOLD \ UNIVERSITY/ INDUSTRY COMMITTEE SENIOR TECHNICAL ^ COMMITTEE DR. FROSCH, CHAIRMAN NASA _^ AUTOMATION _ …](https://reader035.vdocuments.mx/reader035/viewer/2022071412/610974a6dd4dbd69785f4e76/html5/thumbnails/11.jpg)
automation. Instead of individual computers, a distributed
but integrated data management system is hypothesized, with
control and monitoring accomplished from the laboratory
module. Each facility could be either replaced entirely
with newer designs over the years, or be upgraded in space.
3.1 GaAs ELECTROEPITAXIAL CRYSTAL PRODUCTION AND WAFER
MANUFACTURING FACILITY DESCRIPTION
The conceptual design of this facility conforms to
design requirements which were developed as follows:
The projected demand for GaAs microelectronics of the
quality attainable in the space environment was defined.
The results are presented in Figure 3.1-1. Microgravity
Research Associates (MRA) provided the reference data for
this figure based on their own conservative marketing
research study. Because of the possibility for development
of other materials and/or processes, a saturation of demand
is conjectured. If, as MRA predicts, demand increases, a
second, upgraded system can be added.
MRAPREDICTION "
2.000
£ 1.000
100.000
50.000 F' 5
ti
20,000
90 92 94 96YEAR
2.000
ASSUMPTION
INITIAL AUTOMATEDFACILITY SHOULDPRODUCE 20.000WAFERSPER YEAR WITH GROWTHPOTENTIAL TO 80.000 W/YR
Figure 3.1-1. GaAs Projected Demand
-9-
![Page 12: GENERAL ®ELECTRIC...V. ANSELMO 1 STUDY MANAGEMENT CAL SPACE J. ARNOLD \ UNIVERSITY/ INDUSTRY COMMITTEE SENIOR TECHNICAL ^ COMMITTEE DR. FROSCH, CHAIRMAN NASA _^ AUTOMATION _ …](https://reader035.vdocuments.mx/reader035/viewer/2022071412/610974a6dd4dbd69785f4e76/html5/thumbnails/12.jpg)
The size of the furnace and size of the ingots to be
produced were determined by integrated analyses. An ingot
diameter of 3 inches was chosen primarily because furnace
power is proportional to area. Five inch diameter ingots
are expected to be the earth-based industry norm in the
near future, but require nearly three times the power
required for producing three inch ingots for the size
facility projected. This would be prohibitively high for
the IOC space station. Also, because GaAs is extremely
fragile, automated handling of three inch wafers will be
less risky.
Based on the 3 inch size ingot and the projected demand
curve, the furnace was sized to meet a startup capacity in
the 1995 time frame of 20,000 wafers/year at 26%
utilization, and require one third of available space
station power currently planned for the Initial Operational
Capability (IOC). Eventual growth to 95% capacity (near
continuous operation) would produce 80,000 wafers per
year: this would require roughly one third of the
projected space station power available by the year 2000,
and an optimization of the automation system into one which
is almost fully autonomous, including servicing and
maintenance functions.
A power recovery system for the furnace is incorporated
in the conceptual design. Further study would determine
more precisely how much net power would be required.
Analyses should also be accomplished to determine if a
separate power source would be warranted for each facility,
and to determine if other power reducing techniques (i.e.,
pulsed power) can be effectively employed.
A timeline study determined that each ingot should be
grown to a thickness which would yield three wafers per
-10-
![Page 13: GENERAL ®ELECTRIC...V. ANSELMO 1 STUDY MANAGEMENT CAL SPACE J. ARNOLD \ UNIVERSITY/ INDUSTRY COMMITTEE SENIOR TECHNICAL ^ COMMITTEE DR. FROSCH, CHAIRMAN NASA _^ AUTOMATION _ …](https://reader035.vdocuments.mx/reader035/viewer/2022071412/610974a6dd4dbd69785f4e76/html5/thumbnails/13.jpg)
ingot. This is because energy required increases with
ingot thickness, the source crystal can be better utilized
for this size growth, and adequate time is allowed for tray
refurbishment.
Figure 3.1-2 is a block diagram of the facility. It
defines the elements and automation processes required. A
conceptual design was ' accomplished for each element and
automation process, and packaged into a standard fourteen
foot diameter module, as shown in Figure 3.1-3.
FURNACESUBSYSTEM
Q»^ f ^+ 1
ENERGYRECOVERYSUBSYSTEM
nUNPRESSURIZED
PRESSURIZED
T =TRAYR = RECEIVERI = INGOT (=S+P)S "SEEDP =PRODUCTCj" CARRIER 1C2= CARRIER 2D = DOLLY
h-0-
Figure 3.1-2, Crystal Production And Wafer Manufacturing
Facility Block Diagram
-11-
![Page 14: GENERAL ®ELECTRIC...V. ANSELMO 1 STUDY MANAGEMENT CAL SPACE J. ARNOLD \ UNIVERSITY/ INDUSTRY COMMITTEE SENIOR TECHNICAL ^ COMMITTEE DR. FROSCH, CHAIRMAN NASA _^ AUTOMATION _ …](https://reader035.vdocuments.mx/reader035/viewer/2022071412/610974a6dd4dbd69785f4e76/html5/thumbnails/14.jpg)
UJ
UJ
UJ
0)4-1(0
•o
4JU
TJoM
m0)>MU
co
(VM3tn
-12-
![Page 15: GENERAL ®ELECTRIC...V. ANSELMO 1 STUDY MANAGEMENT CAL SPACE J. ARNOLD \ UNIVERSITY/ INDUSTRY COMMITTEE SENIOR TECHNICAL ^ COMMITTEE DR. FROSCH, CHAIRMAN NASA _^ AUTOMATION _ …](https://reader035.vdocuments.mx/reader035/viewer/2022071412/610974a6dd4dbd69785f4e76/html5/thumbnails/15.jpg)
3.2 CRYSTAL PRODUCTION AND WAFER MANUFACTURING FACILITY
AUTOMATION REQUIREMENTS'-•• -• *'•" MtlZtoi• •••• --'-JO &003 ?,
The details of each element and automation process are
contained in Volume II. Automation requirements are
summarized in Figure 3.2-1.
Much of the automation is in the form of process
mechanization schemes similar to those used in factories
today for materials handling and manufacturing. However and
robots are conceptualized for servicing and maintenance
functions, primarily because of their flexibility. Their
operating profiles and timelines can be easily altered or
upgraded by software, and the configuration of end
effectors and their functions can be easily changed for the
required applications. In the concepts presented herein,
these teleoperators and robots are assigned the added task
of materials handling. if it were not for the challenges
of maintenance and repair presented by the limitations of
the space configuration, the materials handling could be
accomplished by straight-forward process mechanization as
in earth-based factories.
Artificial Intelligence (AI) will play an increasing
role in the operation of this facility, primarily in the
areas of process planning and control, and maintenance.
The complexities of electroepitaxial crystal growth, and
the sophistication of the furnace and slicing equipment
warrant an integrated "expert" . Expert process and
maintenance controllers offer an expanded knowledge-base to
aid Space Station crew members in identifying,
troubleshooting and handling anomalies in the crystal
growth process and associated equipment performance. As
conceptualized, the expert controllers would perform the
following functions:
-13-
![Page 16: GENERAL ®ELECTRIC...V. ANSELMO 1 STUDY MANAGEMENT CAL SPACE J. ARNOLD \ UNIVERSITY/ INDUSTRY COMMITTEE SENIOR TECHNICAL ^ COMMITTEE DR. FROSCH, CHAIRMAN NASA _^ AUTOMATION _ …](https://reader035.vdocuments.mx/reader035/viewer/2022071412/610974a6dd4dbd69785f4e76/html5/thumbnails/16.jpg)
V *•»
S ** at
1 | S§ ^ I
g w •
5 c o •—« u «a m»• UK- CTi *-s-i I sISMf . k M
o. 'Sr*' °r
I s .*• ' • " 'C M v_
I 1 S = 2 B"^ S ~ 8" « e 2 ~o* » "£ ~* "" ^ c * *~
a>44 at "*Sfei .— 4 4 M » —«< ? ** ^ *fc **" "O B Q
£ at * O ^ •« 1 * 5 § t
f" 1 £ HI? I" 5 ~gSIs 1; 8"lg5 g o | |
X *** eZ.3 %- l a t ^ ^ - S ^ O - o C ^S i •*» O « O B « t f * « w » C 4 4 C * a i O o >»- — .0 u IE al »** O — • O o c i> 44 o
5 — -2 ** E" c? -fi 8 '£e>5£^a£ u o"v» c *wt£ 44 u<*. a 5 g * § 44 01 44 ^-
5 o § Z TS «! Z 2 £ f c ' * * 5 § o » o . i * e J 5 o 5 o•• S £ -o a> o 9 *o ^ % 8 ? 8 8 c g ^ « a . e < e E^ 3 *o <a 44 *s u • p a i a i O * 4 9 ^ o ) Q i - a i a t ^ a iS UJ < O •** ^ ^ O £ ™ *•* fli <K ^ l-*VO^" t e fe*f f
-Q < o c c o •* J*{ *• "~~
u. o O O ™ Cl O
ORIGINAL PA6H ISOF POOR QUALITY
N
C 1
at 44u - 44 at v*« -0 c ~ a*
* " l a . » 5 .
o t5.T S S • u *
5 5 £& 3 £ i ||S S o>*. w *• * *^ M i o u o u c f u c ° "* o ^- -k- O <O o OvO > 44 O ^ £ jr C
G L. 3 i; 44 r_ 44^
ge^ »•§ .£ Si C ^5* — u o * o 44 5 T> Ji ' 1? .5^443 — c ^ at ot at « vt ««.
44 M ^ C I M M O U *O
""? - tssa-ors I £«t - c * * * — So «7JS 2 "" — 'S'5 c"o * — ?jil *i jo>£ * C ! M V * > « 5 o-2
•— *u c X0-1 II 1 l<l 1 (S I 1
o o o
MOUVZINVHD3H SS330M
«J CQ O»
3 fefc ss>>s1 sl^ lliie? i« g*.£ "SS£ = , 1•« 3w» • « C V « C O ' > * W
< a>4aa> "~ u2'^£! 'oo t.te Jd'-** 44 tt. — x *" > -- Oft w-o -o >, o> 4-» «*-o v i c e o» at — »«
Q 4^<a*a» -o .c E "O 44 » O -afi! e — O *- 3 i- i- "c 44 44 o»
= ~- ° E »*''• =5 * C
S t&| rlSlr'S8 -1
g "SZ^ C ° O 0 c u 4 ^ 1^
* u t i w - ^•o«J2—S.ib « —
y v t u c o a»S.cZ I J Q ' *> §5
g j§,slil f ss r^Ss 5iy ^ r A r
<. -x O < t-•o -o 44 a> ^- trt
O o « •<= >- c_ _ U 44 3 -D idc *« — O n at3 W = * *« O —
o *- — wu <k. 3 at at i- c-*
^ ^ e e — p ^ ^«n u o u • a. oE ID —
f. >t+* k. 3 c at •%. O OS Cu *• »-£ g - tt
i ti5 s? si s i § §
s ^||•e at c
o 44 -o &»£ e•c^,g e o»*
— \_ £ Uvt 44 c a - atc a> -o o a-
5 v^I"
3 -0 "• > «
at at at*o fi44 jr e Q. > e|r.8-=si-SEs<o o » —*Q. U C • =O» « O T3 O v»
*~ >»" 043 —
wt — at — u. oa* */t E a. a> .0f <O — 4) CX O
*~ *V "*
O
§ 0 |
u.
5 ou u<a a.
<a <aO « e u o ° * - ' ' g w a i a. ^- at at
UM << -O JD k. «M •* — O >* C « UJ C 4 4 4 4 . O. « •«f c 2 10 c^- i _ o - - u e -J ^ * » 4 4 g a t ^ M
M f» H » O < O k ^ ' 9 <OW E V- 3 *— 44 »•» . ** ,^ Q c •^^ v> H- c a> a> %• « it < wt j
5 ^ To vo > * • wS S E • 5 S Svua tk .O c c • — »- -*^ a> — o> ^ e v* io«ua*i
v r c— j *O 'o^a t * * ' J c *- o v* « a» % e o w »»-O o *o <ou O t ™^ a» *» 44 v» t/> * ^ v» — w»M O k.a*«ta iO«> c"~ 3 > t w a * * O M « M »-• i- c c *• c ~J S U > > < - - < K > ^ 44 »o c »j •»>><• .j *— ri * «J w c££§*>»x^aa^a :4^ccowa: S *i -01^ 44^
S 5 2. i i t • i i c5 11 g gS i i2o O £*O Jr t i i 0
n- at at-g < . ^u . *•M — m < <o *o v* -o v>in 44 »_ e E c E
* 9 vi at <o vt *j a> vs** c u ^44 a*5*** a. "5 a. S ? « > >, i
> — c *O *o 44 at at at 3 •M u v* a. c c oc E ex «t b-— at —trt 1 1 *c X 1 1 t
0 X 0
SOI1090H
=•a . x
ir.l 5r°||>. 1.C Vt U — U
§ e ai 3 o» •3 O>H- • 44 44
a» — «•- "O u
4?zi?l^s ZS O..C ^ 44 44 C
o. o c c atct <o o «o a* O c S
E u « a.v»— -p^j- a. coS^I°°3 Zlu U C « O C >
2 1; o - s § s-o^sr^^g 6T.«j c a o ^ — 3 5 » Jo
.Z- ^^
"o1-
1 <^ 8-00 0 £C U-l•O vt C C
vt o O at
J O «0 * 3 —l_ 44 U O 44a. (/> at i- u
Q. 1- «Cat •— v Ou <e v» ae -*«-<o — o» uj — • vt oc u u _* « 44^ se d i =s1 *- s 1 "144 "ie 44 o> at 44
o c c u c ui <e E vtuj Q •» Q O ^ o. 4»_j O > O z u. — coO - O V - O S T9 §•$.a- c i_ at c uj c ujo^* « 3 44 <O 44 K- « 44 —
z 44 ,0 CK c a t *3 u <o c i. at ^ w a > 4 4 >O O *- — O 6 < O E « «
44 at "D 44 a. X 44 o.— oI/I — Q. U — — — — 3M C E O C 3 1- C 3 M - 0uj o v q o o * ce o v c cO X^-OXUJ ui X Ui ^ <0
8 xa. o 0 O u o O
33N39I1131NI IVIDUIlilV
C"»•
VI^
-po(0
•o
4Jo
O
CM
frl
0)
•H
-14-
![Page 17: GENERAL ®ELECTRIC...V. ANSELMO 1 STUDY MANAGEMENT CAL SPACE J. ARNOLD \ UNIVERSITY/ INDUSTRY COMMITTEE SENIOR TECHNICAL ^ COMMITTEE DR. FROSCH, CHAIRMAN NASA _^ AUTOMATION _ …](https://reader035.vdocuments.mx/reader035/viewer/2022071412/610974a6dd4dbd69785f4e76/html5/thumbnails/17.jpg)
(1) Process Control - The expert process controller
interprets assimilated data from process sensors.
This information is evaluated against the
knowledge base with subsequent diagnosis, and
corrective action derived for identified process
anomalies. As an example, the disruption of power
to ' the furnace during the growth cycle will
require a process planning decision to determine:
o Which furnace cells, if any, should be
cleared, and plan for the recycling of
source crystals, and discarding of waste.
o Optimum schedule to provide ingots which
can yield one or two wafers, instead of
three, and complete, the wafer processing.
(2) Maintenance Control - The expert maintenance
controller receives inputs from the furnace and
slicing/polishing equipment monitors. As
equipment anomalies are interpreted, the
controller then flags abnormal operation before
hard failure occurs. As the expert system evolves
it will ultimately isolate the equipment fault,
diagnose the cause, and indicate methods of
handling the function.
-15-
![Page 18: GENERAL ®ELECTRIC...V. ANSELMO 1 STUDY MANAGEMENT CAL SPACE J. ARNOLD \ UNIVERSITY/ INDUSTRY COMMITTEE SENIOR TECHNICAL ^ COMMITTEE DR. FROSCH, CHAIRMAN NASA _^ AUTOMATION _ …](https://reader035.vdocuments.mx/reader035/viewer/2022071412/610974a6dd4dbd69785f4e76/html5/thumbnails/18.jpg)
3.3 MICROELECTRONICS CHIP PROCESSING FACILITY DESCRIPTION
Design requirements for this facility are based on the
seven mask process for GaAs Very Large Scale Integrated
microelectronics manufacturing developed by General
Electric. This is a multi-step process which starts with a
polished wafer and ends with one which requires only
packaging functions which are more cost-effectively
performed back on earth, as shown in Figure 3.3-1.
1 1 1
f-MAHOMfcrmin 1 I 1
IPFATT1M ttltlATlM
ITruAHoiRinviin
J. CNMMUMAM |. StWKSfBIUttMUX %. W«TKT«A» It. ttftTt h CCMOTTKY •*• L.J?1*1 "' l 1**1100* ** 2?™'*1
, ETCNPUUOK •. •murrs»*t ti. vomiiA«0^* 11 wvmn it ircmtmnAi 11 cmvuRIGISHIATIW I. OEPOKT CAf WJOLATOH f«OMB* t(MT*CT TWt/A« It. OCPOSinO VHOOn
, kMTLAMTII-l I MnAtnVUUnt 11. ALLOT CWTACT1 IKSUUTO* II. VOTTIR1
Figure 3.3-1. GaAs VLSI Fabrication Schematic 7 Mask
Process
This process is similar to that currently used for
silicon chip manufacturing, but requires fewer steps. The
current trend is toward totally dry processes, which is by
far the easiest way for implementing such a facility in
space, because any process involving management of fluids
would prove very difficult in the microgravity and high
vacuum environment.
-16-
![Page 19: GENERAL ®ELECTRIC...V. ANSELMO 1 STUDY MANAGEMENT CAL SPACE J. ARNOLD \ UNIVERSITY/ INDUSTRY COMMITTEE SENIOR TECHNICAL ^ COMMITTEE DR. FROSCH, CHAIRMAN NASA _^ AUTOMATION _ …](https://reader035.vdocuments.mx/reader035/viewer/2022071412/610974a6dd4dbd69785f4e76/html5/thumbnails/19.jpg)
Seven subsystems are required to manufacture GaAs
chips. These subsystems are diagrammed in Figure 3.3-2,
which also defines the individual steps of the process flow
in sequence and an estimate of the time required by each
subsystem for each step. Note that many passes are
required through the E-Beam Direct Writer, while only two
or three passes are required for other subsystems.
Presuming that only one of each of the subsystems is
utilized in the facility design, the automated handling of
wafers requires versatile teleoperators and/or robotics,
and complex scheduling to accomplish an efficient
processing rate.
21
PLASMADEPOSITIONSUBSYSTEM
1.2 MIN
(A)
10
22
IONIMPLANTSUBSYSTEM
(E)
.2 MIN
(B)
PHOTORESISTSUBSYSTEM
1.2 MIN
11 15 18
RAPIDANNEALSUBSYSTEM
(F)
.3 MIN
14
26
1216192327
.. 5 (D)REACTIVEION ETCHSUBSYSTEM
1.2 MIN
20 24 28
(0E'BEAMDIRECT WRITESUBSYSTEM
2.0 MIN
13
(G)SPUTTERSUBSYSTEM
.7 MIN
17
I(H)
INSPECTIONSTATION
1.0 MIN
I
29
25
Figure 3.3-2. Microelectronics Chip Processing Flow
-17-
![Page 20: GENERAL ®ELECTRIC...V. ANSELMO 1 STUDY MANAGEMENT CAL SPACE J. ARNOLD \ UNIVERSITY/ INDUSTRY COMMITTEE SENIOR TECHNICAL ^ COMMITTEE DR. FROSCH, CHAIRMAN NASA _^ AUTOMATION _ …](https://reader035.vdocuments.mx/reader035/viewer/2022071412/610974a6dd4dbd69785f4e76/html5/thumbnails/20.jpg)
All subsystems used in the conceptual design existtoday or are in;an advanced state of development, but they
are in earth-based configurations. Each is made by several
manufacturers and usually are delivered and incorporated
into microelectronics assembly lines as separate units.
Each has their own command and control hardware and
software, vacuum chambers, and raw material and waste
material handling equipment.
Wafers are generally passed manually in cassettes
between the subsystems in clean rooms designed to filter
air to the Class 100 to 10 level. Even with this level of
cleanliness, contamination is a problem with silicon based
chips: GaAs chips will be even more vulnerable, especially
if chip density increases tenfold or more as expected.
Several manufacturers, including VARIAN, are developing
more fully automated assembly lines incorporating robots
and conveyer systems to replace people in the transfer of
cassettes from, one subsystem to another in an earth based
environment, but the vacuum management is still a major
hurdle in achieving full automation. This problem would be
overcome in the space environment, where a full vacuum
facility is possible.
The GE concept for this facility is therefore one which
is based on the state-of-the-art subsystems, each without
complex vacuum equipment and individualized control
hardware and software. Working meetings with each vendor
resulted in a repackaging of each subsystem and integration
into a fully automated space facility as shown in Figure
3.3-3.
-18-
![Page 21: GENERAL ®ELECTRIC...V. ANSELMO 1 STUDY MANAGEMENT CAL SPACE J. ARNOLD \ UNIVERSITY/ INDUSTRY COMMITTEE SENIOR TECHNICAL ^ COMMITTEE DR. FROSCH, CHAIRMAN NASA _^ AUTOMATION _ …](https://reader035.vdocuments.mx/reader035/viewer/2022071412/610974a6dd4dbd69785f4e76/html5/thumbnails/21.jpg)
43O
COo
•rHcoM4JO0)iH<DOMO•H2
wf*4
(0u•o(V4Jme04Jo<:
•
jjo(Vocou>-P
•H^J^^
•H
On>faenc
•!-(
(0to0)ooMcu
n
(V
en•H
fa
-19-
![Page 22: GENERAL ®ELECTRIC...V. ANSELMO 1 STUDY MANAGEMENT CAL SPACE J. ARNOLD \ UNIVERSITY/ INDUSTRY COMMITTEE SENIOR TECHNICAL ^ COMMITTEE DR. FROSCH, CHAIRMAN NASA _^ AUTOMATION _ …](https://reader035.vdocuments.mx/reader035/viewer/2022071412/610974a6dd4dbd69785f4e76/html5/thumbnails/22.jpg)
Robotics are used to transport wafers in cassettes to
each subsystem in the desired sequence. Control is highly
automated, but supervised and occasionally monitored on the
ground and in a pressurized laboratory module by a space
station crewman. A distributed, fault-tolerant data system
is required to manage each subsystem and the robotics and
other material handling mechanization.
Crew access is provided for repairs and maintenance as
a starting point for this concept. After use in space, the
facility should mature, where nearly all repairs, servicing
and maintenance would be accomplished through teleoperators
and robotics remotely controlled by the software executive
controller and the on-board crew.
-20-
![Page 23: GENERAL ®ELECTRIC...V. ANSELMO 1 STUDY MANAGEMENT CAL SPACE J. ARNOLD \ UNIVERSITY/ INDUSTRY COMMITTEE SENIOR TECHNICAL ^ COMMITTEE DR. FROSCH, CHAIRMAN NASA _^ AUTOMATION _ …](https://reader035.vdocuments.mx/reader035/viewer/2022071412/610974a6dd4dbd69785f4e76/html5/thumbnails/23.jpg)
.3.4 GaAs MICROELECTRONICS CHIP PROCESSING .FACILITY
_.-,... AUTOMATION REQUIREMENTS . .. - .,. ; , - ; I . . , • , . , ; . „ ; : ,,
' • • _ • " ' .-' . • , • • ' ! •-.'•'• * * ": -^ '".. ' •• , ' ' ' • ' " . • ' ' .'.'•) V-! '; '• ' .'
...., ? Automation. requirements, are summarized :.In Figure
.3...4-1. Details are contained, in Volume II. , ; _ ,
In addition to ...the . state-of-the-art mechanizatipn
required for each subsystem process, robots are
conceptualized - for transfer' of : wafer, cassettes .between
subsystems.. These robots, would . be programmed, to-work
together, to move the wafers from, sub system,-.-to subsystem
quickly and efficiently, ;.and to perform cert.ain servicing
and maintenance functions as well. This requires automatic... - . • • 'V -• .^ . . -. .-• i ' •' • • J. , . . .•- - . .
change out of end-effectors, and reprogramming of. operating
profiles.
Software and sensors will be used to control the
process most efficiently by optimizing robot movements to
prevent interference between robots. For example, the
executive controller must know where all arms of each robot
are located, its motion, and its current task, and
coordinate activities between them.
Each robot is equipped with both vision systems and
tactile sensors to precisely locate input and output
devices for the subsystems, to retrieve stray containers or
fragile wafers, and to perform maintenance functions while
the other accomplishes routine processing.
Artificial Intelligence (AI) will be used to identify,
troubleshoot, and handle anomalies associated with
subsystem processing and maintenance. The complexities of
chip fabrication, coupled with the uncertainties associated
with space manufacture give rise to the need for an
integrated "expert" system. Expert process and maintenance
![Page 24: GENERAL ®ELECTRIC...V. ANSELMO 1 STUDY MANAGEMENT CAL SPACE J. ARNOLD \ UNIVERSITY/ INDUSTRY COMMITTEE SENIOR TECHNICAL ^ COMMITTEE DR. FROSCH, CHAIRMAN NASA _^ AUTOMATION _ …](https://reader035.vdocuments.mx/reader035/viewer/2022071412/610974a6dd4dbd69785f4e76/html5/thumbnails/24.jpg)
u
I
~ - T> a|| -S fe S^SJSr t t ? Is ~S*SS* - 52is*?S § .-1«^J"iS • 5^
^S^o — ? * ^^* «9 N « U t/1 Dl
u « ^ O C V — — •§ 0 « — « 4- O— *<stj; fe ts £
5 S g § " S£^ _•> i». • -o O *- *o — — E c oi— oO « « O 3 «•• vt 4v E —
^ti^S £ 1 SS£ 5
£ t £ Q.C 5z A:55 ^Ij £ £ «f 2 "5 •• ••
z >4* UI
? = * £
5 1 |» * e *°
>-2 - S z >— 4V UI C UI O
vt *" in M J ' M U M ^ S M iZ
| ;„ g t"?f ^ || S SaT o *"» in — > uj o -o c? vt at 1
S— • — uj vi — . ce u at z vt * *4-* W t- * 4v 4, 1-1 Ot 6 CX • Q O> O ki
•-• E — O oi t- — j* o *_ «o « 4- i- ua.i— "°*~J5 S" «3 *" 8 * "o « • ex— ui -o O i
ORIGINAL PAGE -ISOF POOR QUALITY
c*1 1
CO >
iS E S '•d C vt i
SO o» o» s O v*4V OJ Cg— VI 4V «
**" S 4^«t- "« S U*5vt at h-t 01 v u vt o <o 4^
Z « i_ wtnoio u i— v»
> irt v» <_> » O — 4v u 1-0*01|>- vi — i>S 4v c O z oi-oa*
CD U £*• C C — Vt »^ t^iSlJI 1} V O * O O O * C X O t A j/t "O3 —1 Vt MWU ^^ « B O >k >- •> Q.- —
zUI
in >» M
CO Ol C3 e ^ v>UJ • .
•^ *~ ^ « vi iS .c z> w c c ?Z 1- V» id O Ot —
M ^«- P «_«.'u«t £T9f-0 < -0 *J « C U vt* in * P « x » > — o> >»
1 l^f 1 ll!Lt| " Us IR f lees 1 Istfleatl =* IsslI 1 lie § Illifef?1 III 1 Illll § III II IIII - Illlsll l £ Illll s 111 i Ilfclfl^ J0> -.o-^oo. y _.«o. g j oo.
5 i? a. g
KOUV7INYHD3H SS3DO«d
_
¥ 5 ?S ,§•« |
< e u t- o *
>- —4- 4- Ol
1 flssl
~ -f-f«l8 o^55 g
< C i. O> C
S ££~~£
5 s^I i§(J O *•» vt —< .0 w w vi >o. o 4- o <e cl/J Of V>«- U 01
f
£ a*
S 0
a. A.
i S
5«s
in in vt w £
O ui « ai
O *n ^ t:
2 * 5°l
^ •
M ^ •._4- Vt <O ' «— ^ C*ldO >> « — c — ex 4v at « —-O 4V *v — O ^ E C E C 4 V
tt ~E O at c -^ ? - 4 v v i A o E O 4 v ' c— - — M O U 4V »j Ol fUC —K 4-> vi — .0 V t C C I - 4 v > ,
4V CX Vt >— -O — CX 4v g 4V 'i_ £>»uai . o at «« O v* 4V O -^ CX — O C 4-* t- <«C X 4 V V I O U - M O C ° S *
;s gg.-01 U *-
;>. i"-g1* 4V 4v «i v
§ 4V CX Ot £ ^ -Q e < * 9 <— i. 4V w — 3 • C
0 VI -O 3 — O • g C *"* ~~
•— «* <B C E * "O<e vi a> c vi 4v ^ e c -— vt o E «* — -o & <oi. or e O — a* 3 « oi — £. •
« 4v vi *v 5,_ >, O at O~W Vt U C ^ vt IQ ^ %-
a* E £ — o. c « co i v t —
§ O «» c •— "u c u o <eU « vt «v — W 01 O CX
w « O — 4V 5 W Z U«O.U
4V«
1§ W 4
i? J!l:*5 *"»*' u= — c *a 91 3 V. O 4V
' 5 °" c 4v c
3 ** * — * Si
- "~ ^o ex"-vi — O Oi0 .kl — — Ol 4va. v* — ex— c
:a ^ssr^
n
_| S cU M (-1
CO Z vt >» CX4v vi
5 ° §^ "1 3..01 -0 t 4v £ S. TJv * a t a.«v Ov> is ~ r*
5 * S— *" 2 0 ^^^>l <« i 4v O 4v Ma O W* 4V -0 S U "-
— 1 Ol « OlC '^ Ol VI — ••o 4v vt « oe — h. u.at O 4v- vi o ui «— t: uj * 2 c 3 S 5 "°"* 'u de « < — T» < uj co u. x* 01 c o <-> 4v ot c a.4v at O o. "Q « 4 v - o o O » o> atS 4-* — f/! f t * ) 4 - t c i l O — UI U 0 U
4v 4V Je OO CO Ot O 4v J Id M} A
vt vt 4v if c "e c v* c 4v t— ex cx cx•• vi in i— «BO oc3«/i at * «
S o** z "e ^'"'cv. lj ""-»?«* o oiO e x w o i O w * z 01 • 01 ot
t/» o cZ a. o_ cc i/t ui «-t er to a. z oe oc oe oe i
. * . . . i ....
i
V> VI
*6 e?
c <o0> 0>
jLcUI •—
vi "O
:l' O «*— t. c>- 0. =D >i
O) 0 W Vta. 3 o M
> TJ*- •
S585S
S3I109W)
M VI
— — 4V c W CX• X « CX X— »£ E *
— 4V 01 -0
f — -o c c* « «at• « c *
"1^1 5
C C tf— O)
g^S-SS"? .— Ot— U — VI
U 0 >» C *-
Z -^«g^o** B -^43 — *o B
Ol f VI 4V i_ oi
— 01 vt *ot ai c v*a. .c >**- o £ O >iUI 4v VI O— 4 V O V t
f
v>
vi -O CVI Ol *
S * 4-O CO viW <w 3a. *-**^ e 3r 2 v.
S*^ v> » t- Oa. > ix —
ol - ' w o u c u »<UI O — Ol Ol * 4Vv 4v 4v CX— 1- VI %.
— « VI 0 U W O
— i ac Q* 3 **
S VI — *- !« *^ *. Ot— Ol 4V vt VI Vt O OZ U Cl VI O! 4V —
8 k O . O l O O I ' O 4 V v t
wi c « a. «- z oi CD(/I Ot«-i Z UI VUI 4V «v ^«_ «CJ « - >> oi >< id ui c
s =cs?i?sl»— VI C •» 4v Ol >4v
OC M O f O C V C O i O lUI « I/I K4 l-l «-• O CC O
a.
o*• « 0
** U — o -C l-
£ •— O cx
• 5 c
d t- * O «* C
Sz^s si8^ ^.3
- o *- er
•§!« £g2«&^ "?
IIIE= "1= 2'™ * & e °-==5-5 B >«< 5 3 « a •* c
— N
« —10 At•O 4rfg M
O « *- *
" " "Z 8o S S - S§*v -o * C * u
e v> « & •- W- i : sis isS I « 5£S "tS S *- uO •<"~ " J I**" *•»-j £ i -c- s°^4^* p vec 3
££ bj *n^f .a^v (J vtv E*'.
"•"°c S e.o? ^™*
ill!1 i ii! ii-
33N39PinNI 1VIDUUHV
(0CM
cCOu
O
(UM301•H
rc
0)
(U
(0
-22-
![Page 25: GENERAL ®ELECTRIC...V. ANSELMO 1 STUDY MANAGEMENT CAL SPACE J. ARNOLD \ UNIVERSITY/ INDUSTRY COMMITTEE SENIOR TECHNICAL ^ COMMITTEE DR. FROSCH, CHAIRMAN NASA _^ AUTOMATION _ …](https://reader035.vdocuments.mx/reader035/viewer/2022071412/610974a6dd4dbd69785f4e76/html5/thumbnails/25.jpg)
controllers offer a knowledge-base from which the SpaceStation operator can draw detailed explanations to
implement timely process adjustments and equipment repair.
As envisioned, the expert controllers would perform the
following functions:
(1) Process Control - The process controller
assimilates specific online data from subsystem
process sensors and inspection probers, and in
turn interprets process anomalies and generates
appropriate responses. Often the generation of
appropriate responses requires simulation of the
potential effects of suggested corrective
actions. This differs from conventional computer
controlled feed back systems in that it can
respond to complex situations by applying domain
expertise to diagnose and correct deficiencies.
The process controller will also make planning and
scheduling decisions as process adjustments are
implemented.
(2) Maintenance Control - The maintenance controller
assimilates real time data from subsystem
equipment and consumable monitors to interpret
equipment anomalies. As anomalies are identified
the expert controller can implement maintenance
tests, note possible deviancies and flag abnormal
transient operation prior to hard failure. The
expert system can isolate a fault, diagnose the
cause, and suggest or implement corrective repair.
Periodically, an electrical probe test will be
performed on selected finished wafers. This consists of
performing up to 73 electrical measurements at various
points on a selected wafer to determine quality and process
accuracy. The process will be accomplished automatically
as it is now, however artificial intelligence techniques
need to be applied to efficiently determine the cause of
any anomalies and appropriate corrective actions.
-23-
![Page 26: GENERAL ®ELECTRIC...V. ANSELMO 1 STUDY MANAGEMENT CAL SPACE J. ARNOLD \ UNIVERSITY/ INDUSTRY COMMITTEE SENIOR TECHNICAL ^ COMMITTEE DR. FROSCH, CHAIRMAN NASA _^ AUTOMATION _ …](https://reader035.vdocuments.mx/reader035/viewer/2022071412/610974a6dd4dbd69785f4e76/html5/thumbnails/26.jpg)
4.0 CONCLUSIONS
The two manufacturing concepts developed in this study
represent innovative, technologically advanced
manufacturing schemes. The concepts were selected to
facilitate an in-depth analysis of manufacturing automation
requirements in the form of process mechanization,
teleoperation and robotics, and artificial intelligence.
While the cost-effectiveness of these facilities has not
been analyzed as part of this study^ both appear entirely
feasible for the year 2000 timeframe. The growing demand
for high quality gallium arsenide microelectronics may
warrant the ventures.
The evolution of enabling technologies for space
manufacturing will require detailed planning, and
coordination with the design team. To facilitate the
generation of a responsive automation plan, a list of
Generic Space Manufacturing Activities was developed from
the McDonnell Douglas Generic Space Activities list (see
THURIS Report for activity definitions). This list, as it
appears in Figure 4.0-1, was further developed to reflect
the degree of automation and associated technology
requirements necessary to perform each of the activities
over four time phased periods. The figure accurately
represents the intimate involvement of man in the process
loop at IOC, and the subsequent scaling down of man's role
with time to accommodate the ultimate autonomous concept.
Figure 4.0-2 depicts the evolution of automation
technologies for space manufacturing from initial
development studies, through IOC, to the ultimate
autonomous manufacturing facility. This technological
progression enhances the stated Space Station technology
goals of "maintainability, autonomy, long life, human
productivity, evolution, and low life-cycle costs".
-24-
![Page 27: GENERAL ®ELECTRIC...V. ANSELMO 1 STUDY MANAGEMENT CAL SPACE J. ARNOLD \ UNIVERSITY/ INDUSTRY COMMITTEE SENIOR TECHNICAL ^ COMMITTEE DR. FROSCH, CHAIRMAN NASA _^ AUTOMATION _ …](https://reader035.vdocuments.mx/reader035/viewer/2022071412/610974a6dd4dbd69785f4e76/html5/thumbnails/27.jpg)
a <
z o C3 < 5
I I I I I IOk
MUJooozu12o
K Oo zz «
uco I— >
I I Icc
a oc xn < ui
I I I
< _J co _
inoCM
UJ
o
<U
512
ceoOKUI90<
<0ZUJUJI-
u t-
OCJ
g5o«J 2ES=*H£0o<
a — —
a. > o.-ev< a.Q.o.Q.Q.o.a.<o. cu5" aTeCaTo. tC> < » a. eC> a.-> > cC> >X —~X X * X X X XX X X ;E X — X —-X X X X X —~2 --—-XX —'X —-—"X —- —UI<UJUllOIUUJUJUJUJIlJUIMUJ<UI<UJUJUJ IU Ul < CO < < UJ UJ <X UJ < < LU < <
ce ee ecEEE
EEx^x'E cc ce oc cc ec cc ce cccccccecccccoccccccc
OQm r» n
> oo co'co co co to c/j' ' ' • r- H-1— I
occccc
CO
'o
Ulu
>_J<
— COg g; E KHgs
CO
c
o
CTO)o:
to
•oa>
cu
COat
O
enc
S-3
on3
O)On3Q.
O)
-25-
![Page 28: GENERAL ®ELECTRIC...V. ANSELMO 1 STUDY MANAGEMENT CAL SPACE J. ARNOLD \ UNIVERSITY/ INDUSTRY COMMITTEE SENIOR TECHNICAL ^ COMMITTEE DR. FROSCH, CHAIRMAN NASA _^ AUTOMATION _ …](https://reader035.vdocuments.mx/reader035/viewer/2022071412/610974a6dd4dbd69785f4e76/html5/thumbnails/28.jpg)
ca<o
S
sCM
fc. uJo<85cc
UJ
C9
CD2E
cncn
cn
cn
incn
COcn
CM
cn
cn
cn
<c
ooIo
-26-
![Page 29: GENERAL ®ELECTRIC...V. ANSELMO 1 STUDY MANAGEMENT CAL SPACE J. ARNOLD \ UNIVERSITY/ INDUSTRY COMMITTEE SENIOR TECHNICAL ^ COMMITTEE DR. FROSCH, CHAIRMAN NASA _^ AUTOMATION _ …](https://reader035.vdocuments.mx/reader035/viewer/2022071412/610974a6dd4dbd69785f4e76/html5/thumbnails/29.jpg)
Additional work must be accomplished to develop highly
automated facilities such as those described beyond the
conceptual stage. Such automated equipment is essential
for cost-effective space manufacturing.
Very versatile industrial robots are in extensive use
today. Those conceptualized for use in space will be of a
very different design. They must be able to operate in a
hostile environment of hard vacuum with potentially high
thermal gradients and radiation. While micr.ogravity allows
their design to be lightweight, different kinematics and
dynamics will exist. Different approaches to actuation
devices and end-effectors must therefore be developed.
While the lack of gravity reduces grip and wrist forces,
gravity can no longer be used as a helper to catch things
or hold them in place. Since the robots must be versatile
enough to handle different materials and various repair and
maintenance functions, a quick change end-effector
replacement system will be required. Many of the complex
maintenance and repair functions will be initially done by
teleoperators; therefore, feedback devices, including
visual and tactile sensors, must be developed well beyond
todays' designs.
As more autonomy is developed, the more reliable,
serviceable, easily repairable, and accurate the equipment
must be. It will be difficult to provide the space station
crew the kind of access, information and resources needed
to adjust or repair highly automated systems in the
confines of a space facility to the degree possible in an
earth-base factory.
The major challenge of space manufacturing is
maintenance and repairs. Without the automation
capabilities to accomplish these functions, manufacturing
in space will be unattainable.
-27-
![Page 30: GENERAL ®ELECTRIC...V. ANSELMO 1 STUDY MANAGEMENT CAL SPACE J. ARNOLD \ UNIVERSITY/ INDUSTRY COMMITTEE SENIOR TECHNICAL ^ COMMITTEE DR. FROSCH, CHAIRMAN NASA _^ AUTOMATION _ …](https://reader035.vdocuments.mx/reader035/viewer/2022071412/610974a6dd4dbd69785f4e76/html5/thumbnails/30.jpg)
Artificial Intelligence can be developed for
manufacturing facilities which will provide efficient
control of troubleshooting, maintenance and corrective
action options. Development of "expert systems" to do the
job even better must await expertise to be gained in
operating the system during development and in space. This
means any program must walk before running by initially
providing crew access where possible. As experience isdeveloped, more hardware and software automation can be
accomplished, thus making space factories more productive
by trading access space for more equipment and materials
storage. The space crew will contribute much to this
evolution, and will supply much of the expertise needed to
develop expert systems for maintenance and repair
automation.
Expert systems are very difficult to develop. A data
base must be developed and expanded as a facility matures.
Therefore more human involvement will be required early in
the evolution of each facility. Elements of an expert
system can be developed individually, but need to be
structured to fit effectively into the total system as it
evolves.
-28-
![Page 31: GENERAL ®ELECTRIC...V. ANSELMO 1 STUDY MANAGEMENT CAL SPACE J. ARNOLD \ UNIVERSITY/ INDUSTRY COMMITTEE SENIOR TECHNICAL ^ COMMITTEE DR. FROSCH, CHAIRMAN NASA _^ AUTOMATION _ …](https://reader035.vdocuments.mx/reader035/viewer/2022071412/610974a6dd4dbd69785f4e76/html5/thumbnails/31.jpg)
5.0 RECOMMENDATIONS
We must "get on with the show": to do so means that
research and development must be accomplished in many areas
and certain space hardware developed and proven.
The following five specific programs are recommended as
a result of this study; they are believed to be essential
for many other space manufacturing applications as well.
(1) Space Manufacturing Concepts Development Study
Several manufacturing design concepts, including those
described in this report, would be more fully developed
to define system requirements, preliminary facility and
automation designs, maintenance and repair scenarios,
space station interfaces, cost-effectiveness, and
evolutionary growth of each through a twenty year
period. The concepts would be chosen to assure maximum
applicability for automation of manufacturing processes
and associated maintenance and repair for all potential
space manufacturing applications.
(2) Space Robotics System Experiment
A general purpose, hybrid robot would be designed for
experimental evaluation in space. A hybrid robot
normally operates under program control with sensory
feedback, but for certain applications can be remotely
controlled as a teleoperator. A modular design will
be considered, so that combinations of different
configurations can be evaluated. Self maintainability,
the capability of one robot to perform maintenance,
repair, and servicing as required for itself or for
another robot, will be explored.
-29-
![Page 32: GENERAL ®ELECTRIC...V. ANSELMO 1 STUDY MANAGEMENT CAL SPACE J. ARNOLD \ UNIVERSITY/ INDUSTRY COMMITTEE SENIOR TECHNICAL ^ COMMITTEE DR. FROSCH, CHAIRMAN NASA _^ AUTOMATION _ …](https://reader035.vdocuments.mx/reader035/viewer/2022071412/610974a6dd4dbd69785f4e76/html5/thumbnails/32.jpg)
Performance and design requirements would be determined
using the Space Manufacturing Concepts Development
Study as the primary reference, and reviewed by an
independent industrial/university committee. After
approval, the design would be fully developed, and an
experimental robot, together with its controller and a
variety of sensor systems, actuators, tools and
end-effectors manufactured, tested and flown as
experiments on the Shuttle by 1990. Experiments would
concentrate on maintenance and repair activities, but
also investigate materials handling tasks.
(3) Materials Management Study
Space-based handling of the various raw materials
required for space manufacturing,and the handling and
disposal of waste products and by-products would be
studied. Gaseous, liquid and solid waste products
would be included and concepts developed for handling
of hazardous, valuable and unstable materials in the
space environment. Servicing schemes for replenishment
and disposal would also be addressed.
(4) Materials Handling Experiments
Experiments in materials handling which are necessary
for a variety of manufacturing applications would be
flown on shuttle in the 1989-1991 time frame. Included
would be experiments in gaseous, liquid, and solids
handling of raw and waste materials and by-products,
selected from results of the Materials Management
Study. Handling of toxic materials necessary for
likely space manufacturing systems and collection of
dust-like particles resulting from slicing and
polishing operations would be candidate experiments.
Some experiments would be integrated with the Space
Robotics System Experiment.
-30-
![Page 33: GENERAL ®ELECTRIC...V. ANSELMO 1 STUDY MANAGEMENT CAL SPACE J. ARNOLD \ UNIVERSITY/ INDUSTRY COMMITTEE SENIOR TECHNICAL ^ COMMITTEE DR. FROSCH, CHAIRMAN NASA _^ AUTOMATION _ …](https://reader035.vdocuments.mx/reader035/viewer/2022071412/610974a6dd4dbd69785f4e76/html5/thumbnails/33.jpg)
(5) Space Manufacturing Artificial Intelligence
Applications Study
A university/industry team would study specific
concepts selected from the Space Manufacturing Design
Concepts Development Study to define conceptual
artificial intelligence system designs for control,
maintenance, troubleshooting, and corrective actions
required to operate the facilities. The data
management system requirements for these AI concepts
would be sized and interfaces with the Space Station
Data Management System (DMS) defined thus providing the
foundation for full development of Space Manufacturing
AI applications. This effort should commence as soon
as possible because of the potential impact on Space
Station DMS architecture.
Space manufacturing activities must be closely
coordinated with other Space Station activities. Impact
assessments need to be conducted during Phase B studies to
ensure compatibility of manufacturing missions with Space
Station operations. Manufacturing interfaces that have
been identified as requiring further evaluation by Phase B
contractors include power, thermal, data handling,
servicing and waste management.
o Power - Further study to determine power
distribution impacts on process scheduling;
scarring study to accommodate power expansion
requirements beyond IOC; assessment of GaAs
crystal growth potential for manufacturing solar
arrays to create additional/independent power
sources.
-31-
![Page 34: GENERAL ®ELECTRIC...V. ANSELMO 1 STUDY MANAGEMENT CAL SPACE J. ARNOLD \ UNIVERSITY/ INDUSTRY COMMITTEE SENIOR TECHNICAL ^ COMMITTEE DR. FROSCH, CHAIRMAN NASA _^ AUTOMATION _ …](https://reader035.vdocuments.mx/reader035/viewer/2022071412/610974a6dd4dbd69785f4e76/html5/thumbnails/34.jpg)
o Thermal - Study to assess the effects of thermal
shadowing on manufacturing processes; trade study
to evaluate centralized vs distributed control of
temperatures; assessment of the economics of
recovering Space Station waste heat energy or
solar energy to help drive furnaces.
o Data Handling - Study to determine required levels
of language, capacities and rates; projection of
data handling requirements for the ultimate
manufacturing concept to allow sufficient scarring
to accommodate evolving Space Station
manufacturing facilities.
o Servicing - Study to assess a universal Space
Station approach for raw materials, gas and fluid
handling, study to determine problems associated
with handling toxic materials/gases.
o Waste Management - Study to resolve waste
collection vs dumping trade-offs; evaluation of a
common waste collection module to be launched from
the Space Station for incineration by the sun.
These studies and experiments will help develop new
technologies required for space manufacturing. The studies
will also stimulate interest in the manufacturing
industries through involvement and understanding of the
benefits of manufacturing in space. With the desire of
American and foreign industries to reap these established
benefits, the future of the Space Station Program will be
assured.
-32-